New Zealand

A quadcopter can capture photos from vantage points unattainable for a human photographer, but teloperating it to a good viewpoint is a non-trivial task. Since humans are good at composing photos, the aim of our research is to leverage deep learning to create a customizable flight controller that can capture photos under the guidance of a human photographer. Our system, the Selfie Drone Stick, allows the user to assign a vantage point to the quadcopter based on the phone&#39;s sensors. The user takes a selfie with the phone once, and the quadcopter autonomously flies to the target viewpoint. The proliferation of open source deep learning models provided us with a large variety of options for the computer vision and flight control systems. This article describes three key innovations required to deploy the models on a real robot: 1) a new architecture for rapid object detection, DUNet; 2) an abstract state representation for transferring learning from simulation to the hardware platform; 3) reward shaping and staging paradigms for training a deep reinforcement learning controller. Without these improvements, we were unable to learn a flight controller that adequately supported the intuitive user interface.

AAMAS 2020

Leveraging Deep Learning Models to Create a Natural Interface for Quadcopter Photography

quadcopter · human-robot interaction · machine learning

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2022

A quadcopter can capture photos from vantage points unattainable for a human photographer, but teloperating it to a good viewpoint is a non-trivial task. Since humans are good at composing photos, the aim of our research is to leverage deep learning to create a customizable flight controller that can capture photos under the guidance of a human photographer. Our system, the Selfie Drone Stick, allows the user to assign a vantage point to the quadcopter based on the phone's sensors. The user takes a selfie with the phone once, and the quadcopter autonomously flies to the target viewpoint. The proliferation of open source deep learning models provided us with a large variety of options for the computer vision and flight control systems. This article describes three key innovations required to deploy the models on a real robot: 1) a new architecture for rapid object detection, DUNet; 2) an abstract state representation for transferring learning from simulation to the hardware platform; 3) reward shaping and staging paradigms for training a deep reinforcement learning controller. Without these improvements, we were unable to learn a flight controller that adequately supported the intuitive user interface.

technical paper

AAMAS is the leading scientific conference for research in autonomous agents and multi-agent systems. The AAMAS conference series was initiated in 2002 as the merging of three respected scientific meetings: the International Conference on Multi-Agent Systems (ICMAS), the International Workshop on Agent Theories, Architectures, and Languages (ATAL), and the International Conference on Autonomous Agents (AA). The aim of the joint conference is to provide a single, high-profile, internationally-respected archival forum for scientific research in the theory and practice of autonomous agents and multi-agent systems.

Browse keynotes, discussions, panels and over 300 presentations.


AAMAS is the leading scientific conference for research in autonomous agents and multi-agent systems. The AAMAS conference series was initiated in 2002 as the merging of three respected scientific meetings: the International Conference on Multi-Agent Systems (ICMAS), the International Workshop on Agent Theories, Architectures, and Languages (ATAL), and the International Conference on Autonomous Agents (AA). 

Ethics in Sociotechnical Systems

Live Simulations

Agents are Dead Long Live Agents !

A Multi-Robot Platform for the Autonomous Operation and Maintenance of Offshore Wind Farms

The DALI system aims at making existing conventional traffic control systems autonomous and smart. We achieve this goal by plugging-in a software agent into each existing intersection controller which becomes “the brain” of the controller. The agents analyze the traffic data, communicate with each other directly, and collaborate to execute a timing strategy that improves traffic flow.

DALI: An Agent-Plug-In System to “Smartify” Conventional Traffic Control Systems

Deep Reinforcement Learning for Market Making

This video presentation presents the main work we have done for investigating of Multi-agent Reinforcement Learning for Multi-traffic Light Control.

Integrating independent and centralized multi-agent reinforcement learning for traffic signal network optimization

Agents that are tailored to appear and behave as members of a particular culture are more acceptable by and persuasive to members of that culture than agents that are not tailored. We report a study that systematically unpacks two tailoring components—appearance and argumentation—for virtual exercise coaches designed for the Indian and American cultures. Indian participants who interacted with an agent whose argumentation was tailored to their culture were significantly more satisfied with the agent irrespective of the agent’s appearance.

Downloads

Next from AAMAS 2020

Ethics in Sociotechnical Systems

Similar lecture

Robust Adversarial Reinforcement Learning with Dissipation Inequation Constraint

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAMAS 2020

Ethics in Sociotechnical Systems

Similar lecture

Robust Adversarial Reinforcement Learning with Dissipation Inequation Constraint

Downloads